BELUGA RIVER GAS FIELD GEOCELLULAR MODEL: MODELING A COMPLEX FLUVIAL RESERVOIR

LEVINSON, Rick A.¹, SCHEIHING, Mark H.², DELANEY, Peter³, SCHEEPENS, Claude², SEIFERT, Dirk¹, and BRADEN, John C.¹, (1) ConocoPhillips Alaska, Inc, 700 G Street, Anchorage, AK 99501, [email protected], (2) ConocoPhillips Company, 600 North Dairy Ashford, Houston, TX 77079, (3) ConocoPhillips Comapny, P. O. Box 24750, Doha, Qatar

The Beluga River Gas Field (BRU), in northern Cook Inlet, Alaska, is at a mature stage of development. A program is underway to mitigate the impacts of declining pressure, water breakthrough, and sand production. As part of this effort, a major re-description of the reservoir was undertaken and integrated into a 3D geocellular model. A particular focus of the modeling effort was to identify gas volumes that might not yet be drained with existing perforations. This model has proven to be a useful tool for reservoir management, and it has helped to identify new production opportunities.

The producing intervals at BRU are Neogene-aged, fluvially deposited sandstones of the high net-to-gross Sterling Fm. and underlying low net-to-gross Beluga Fm. The structural and stratigraphic framework of the model was built from 2D seismic structure surfaces and a flow unit zonation based on correlation of major coal zones. Within this framework, major fluvial facies associations were distributed using object-based stochastic modeling techniques. Facies proportions and dimensions were derived from 21 wells, and facies orientations were based upon dipmeter logs and regional depositional trends. Permeability and porosity were then distributed following the trend of facies objects using a stochastic petrophysical model. Water saturation was based on J-functions representing key rock types with multiple free water levels. A total of 10 realizations were created to appraise the uncertainty associated with the stochastic component of the model. From these realizations, a probability cube was calculated that identifies the most likely occurrence of net pay sandstones in the field. A static connectivity analysis was also conducted to estimate gas pore volume connected to existing perforations.

Results indicate that Sterling Fm. sandstones are 99% connected to existing well perforations, while the Beluga Fm. sandstones are only 81% connected. Connected original-gas-in-place in the model is 28% greater than that indicated by p/Z analysis, suggesting that there may be substantial opportunities for well work or new drilling to access isolated pay sands, mainly in the Beluga Fm. This has been tested in two recent workover operations, which have resulted in new pay sands being identified and perforated, leading to increased gas production.